I

The first issue of the Décade philosophique appeared on 29 April 1794. In all, fifty-four volumes of the journal were published between that date and 1807, when, on Napoleon's orders, it was forced to merge with the Mercure français. The Décade was published three times a month (taking its name from its appearance on the tenth day of each month of the French republican calendar) and the periodical soon became one of the intellectual powerhouses of the French republic after Robespierre. But quite what, in this particular setting, an intellectual powerhouse might have been is still an open question. Alongside Immanuel Kant or Jeremy Bentham, and their vast and varied intellectual legacies, the significance of the dozens of writers, including Pierre-Jean-Georges Cabanis, Pierre-Samuel Dupont de Nemours and Jean-Baptiste Say, who contributed to the Décade, is now more difficult to specify. There have, of course, been several fine studies of the Décade and its contributors, notably by Joanna Kitchen and Marc Régaldo, and more broadly by Sergio Moravia, Martin Staum and Cheryl Welch.Footnote ¹ But it is still somewhat easier to associate the periodical with a number of keywords, such as idéologie and science sociale, than with anything comparable to those more comprehensively articulated bodies of thought that came to be labelled “idealism” or “utilitarianism”. “Ideologism” never seems to have existed, and certainly never caught on. But this very indeterminacy may still be an advantage. It may help to open up, both historically and analytically, rather more of the intellectual space once covered by the broad range of subjects and arguments that first helped to shape—and then came to be buried by—idealism and utilitarianism.

The history of both terms, and of the moral or political philosophies that they came to designate, still seems to belong to a number of fairly specialized historiographical traditions, either in the history of philosophy or in the assorted histories of political philosophy, moral theory or international relations theory that once took their cue from nineteenth- or twentieth-century characterizations of the Enlightenment. But, as the old enlightenment–idealism–utilitarianism trinity has slowly disintegrated, all that seems to be left are the names of the initial subjects—ideas and utility—that were incorporated into the later, more comprehensive moral and political systems. This too is an advantage. In different guises, thinking about how to reconcile the spiritual and the material, the moral and the physical, or the honourable and the useful was as old in Europe's history as the Greeks, the Romans or the Christians. This may not be much help in identifying whatever was new or different about ideology. But setting the initial subjects of ideas and utility against the range of established questions with which they had long been associated may be a real help in identifying the continuities and discontinuities involved in this particular round of reconciling the moral with the physical, the rightful with the useful, or the spiritual with the material. What, in the first instance, made this particular round rather different was the question of how the subjects of ideas and utility were connected to the new, eighteenth-century, concept of civilization.Footnote ² Bringing civilization into the picture may make it easier to piece together rather more of the intellectual context in which both idealism and utilitarianism took shape. Thinking about the subject minus the conceptual and material resources implied by the availability of complex ideas or useful things certainly makes the idea of civilization look rather strange. But simply putting them back is not much help in making the concept any the more self-explanatory, since it is not clear what type, or amount, of either the one or the other can give the idea of civilization a recognizable content. It is not hard to see why the question came to matter in France in the late eighteenth century, or why it was unlikely to have had a simple answer. This too may make finding out more about ideology worth doing.

An opportunity to start is now available in the splendid, nine-volume, selection from the Décade that has been compiled and edited by Josiane Boulad-Ayoub with the assistance of Martin Nadeau. Although any selection may be tendentious, users of this edition have been given the clearest possible guidance to what has been included and excluded. The nine volumes have been arranged thematically, with the first volume consisting of a selection of readers' letters; volumes three and four covering the subjects of education and the various public institutions with which it was associated (including the reports of the proceedings of the French Institute that the Décade regularly published); volumes five to seven concentrating on philosophy, morality and politics; and the final two volumes (edited by Martin Nadeau) dealing with the performing arts (spectacles) and public opinion (esprit public). The whole second volume, however, contains all the original tables of contents of each issue of the Décade, making it easy both to see where any of the articles published in the other volumes once belonged and to establish an overview of the wide range of scientific, philosophical, economic, literary, technical and pedagogical subjects that fell under the institute's comprehensive academic remit. In addition, all the selections published in the other eight volumes are accompanied by brief tables of both the main concurrent events and the contents of the issue from which the selection in question has been made. Each entry is also preceded by an indication of the original pagination of, and rubric under which, the article first appeared. This impressively high level of editorial assistance entails a certain amount of repetition (recurrently displaying the same tables over the eight thematic volumes probably adds up to half a volume in itself), but the result is a remarkably well arranged and easily usable edition. It also comes equipped with name indexes and tables of contents at the end of each volume. Although these might have been consolidated into a general index, the only thing that is really missing is an index, or set of indexes, of key terms. However limited or partial these might have been, they would have made it easier to notice connections between the many apparently different subjects discussed in the Décade and, by doing so, might also have helped to reveal the extent to which the journal was either a forum of discussion and debate or, as it is more usually described, simply the organ of the group of writers and thinkers who came to be known as “the ideologues”.

Idéologie was both a term of art and a far from neutral signal. As a term of art, it referred to the study of the origin of ideas and, by extension, of the attributes of the mind, or what happens in thinking. As a signal, however, its meaning is still surprisingly unclear. Its most obvious conceptual antonym was physiology, or the study of the attributes of the body, although this too could include studying the origin of ideas. The overlap raises an initial question about what the point of idéologie might have been. If it was intended to be more than, or different from, physiology, then this raises a further question about what, in the late eighteenth century, physiology was taken to be. Both these questions have yet to be given full answers. Ideology is still usually taken to be a refinement, or extension, of the epistemological theories developed a generation earlier by the abbé Etienne Bonnot de Condillac, who, it is also usually assumed, was primarily concerned with developing and refining the much earlier epistemological theories set out by John Locke in his Essay Concerning Human Understanding. But Condillac's brother, the abbé Gabriel Bonnot de Mably, did not associate Condillac with Locke when he wrote about his brother's philosophical investigations. Instead, he associated him with the German philosopher Gottfried Wilhelm Leibniz. “As for myself”, Mably wrote in one of his late works (written shortly before his death in 1785), “while bemoaning all our miseries, I am persuaded along with Leibniz, that man is as perfect as he can be, since he is made up of two substances as different as the soul and the body.” In this characterization of human nature, and “to know the fate of our captive reason”, he was, he explained a little earlier in the same work, also relying on “my brother's philosophy, which I am simply applying to moral and political matters”.Footnote ³ From Mably's perspective, the mind–body dualism indicated here was one that Leibniz and Condillac shared. As the character “Cleophon” in another of Mably's late works put it, “our soul, hampered by our senses, rises as far as it can to reach its dignity, but, for perfect instruction, one has to go to the trouble of dying” (“Cleophon”, Mably wrote, “was always accompanied by his Leibniz and his Condillac”).Footnote ⁴

Mably's characterization of his brother's philosophy has begun to be corroborated by more recent research.Footnote ⁵ This does not, of course, mean that the line of intellectual descent running from Locke through Condillac to some of the contributors to the Décade philosophique is simply wrong, since Condillac's thought was clearly central to discussions of the nature and purposes of ideology. But the connection between Condillac's thought and whatever ideology was supposed to be may still have to be revised to include a more substantial and durable French engagement with Leibniz's intellectual legacy than has usually been recognized. This, in turn, may have two further effects. The first may be a need to pay more attention to what was going on intellectually beyond the francophone world, and to examine the contents of the Décade as much in terms of the pre-Kantian philosophies of the German-, Italian- and English-speaking worlds as in the more narrow terms supplied by the events and arguments of the French Revolution. The second, related, effect has a bearing on the question of the relationship between ideology and physiology. As is well known, Leibniz's original ontology had a dualistic appearance, with a pre-established harmony regulating the properties of the physical and spiritual monads that he took to be the most basic units of the Creation. But the dualism associated with Leibniz's original system began, in the second half of the eighteenth century, to give way to a more monistic and more strongly spiritually oriented ontology, a development usually associated with the strand of natural philosophy that came to be known as “vitalism”. Here, Leibniz's philosophy came to be associated with rethinking the mind–body problem in terms of an increasingly elaborate (or, in Leibnizian language, more perfect) hierarchy of living beings, with the spiritual predominating incrementally over the physical as the internal organization of bodies became more sophisticated and refined.Footnote ⁶ This has a bearing both on idéologie and on the broader late eighteenth-century French interest in Condillac's philosophy. Going back to Condillac could mean going back to something nearer to Leibniz's original dualism in order to revise and correct more recent vitalist natural philosophy. It could, however, also mean revising and correcting Condillac in the light of the monism of the vitalist natural philosophy with which the posthumous Leibniz was also associated. It is still not clear which of the two objectives idéologie was supposed to meet and it may, of course, have been the case that this was also a subject of disagreement at the time.

One reason for disagreement was the connection of these different versions of natural philosophy to the related subjects of morality and civilization. As Mably's remarks on his brother's philosophy indicate, Condillac's neo-Leibnizian dualism was quite close to Christian orthodoxy, with the senses forming an obstacle to any more perfect human knowledge of what, ultimately, might be right or true or good. Here too it is not clear whether the ideologue interest in Condillac was designed to replace or reinforce this type of dualism. The well-known hostility of the more prominent advocates of ideology towards Kant's noumenal–phenomenal dualism (partly covered in the seventh volume of the Boulad-Ayoub edition but examined more fully in the fine edition of contemporary texts published by François Azouvi and Dominique Bourel in 1991) might suggest the former.Footnote ⁷ But a straightforward antithesis between ideologue monism (or materialism) and Kantian dualism was complicated by the more pronounced, but also more spiritual, monism of Kant's early French admirers. The most able of these was one of Kant's first French expositors, Charles Villers, while another was the dramatist, satirist and republican moralist Louis-Sébastien Mercier. Both, however, had been strong admirers of rather similar versions of vitalist natural philosophy at earlier stages of their lives (Villers had been a mesmerist, while Mercier had been an admirer of the Swiss physiognomist Johann-Caspar Lavater).Footnote ⁸ To them, Kant's critical philosophy complemented vitalist natural philosophy because the picture of a noumenal–phenomenal world that it presented appeared to parallel, or even corroborate, their own interest in explaining how the spiritual gave form and content to the physical and, by extension, in showing how the progress of civilization involved the increasing subordination of the physical to the spiritual. Here, Kant could be used to criticize Condillac. “It is horrifying to see the repeated efforts that have been made to turn conscience and the moral instinct into accidents”, Louis-Sébastien Mercier wrote in a sympathetic outline of Kant's philosophy that he presented to the French Institute in 1802, where idéologie and the type of moral philosophy that he associated with Condillac and his followers were his more immediate targets. “Morality”, Mercier asserted, drawing on Kant's own terminology for the purpose, “is the highest point of our nature and its primordial sentiments, inherent in human nature, exist by virtue of the synthetic unity of the I [Moi]”.Footnote ⁹

Inversely, however, Condillac could be used to criticize Kant. Since, as Antoine Destutt de Tracy emphasized in the preface to his Projet d'éléments d'idéologie of 1801, ideology was “a part of zoology”, it made no claims about anything specifically human in analysing how the faculties of feeling, remembering, judging and willing were able to turn raw sense data into complex ideas.Footnote ¹⁰ This limited brief in explaining what thinking involved ruled out any assumptions either about the special properties of human nature or about the peculiarly human qualities of the categories that, echoing Kant, Louis-Sébastien Mercier associated with synthetic a priori reasoning. At most, Destutt insisted repeatedly, whatever might be distinctively human about complex ideas or abstract concepts was so heavily dependent on the largely unknown properties of human physiology as to rule out further speculation. It was simply impossible to explain why ears heard, eyes saw or hands touched things in the effortlessly differentiated ways that they did. All that could be said was that capacities like these were required for thought, that thought was connected to life, but that life itself was beyond analysis.Footnote ¹¹ From this perspective, ideologue scepticism about the physical paralleled Kant's own scepticism about the metaphysical. However much of a failure it turned out to be, there may, therefore, have been a point to Wilhelm von Humboldt's choice of interlocutors when he presented Kant's philosophy to a mainly baffled ideologue audience in Paris in 1798.Footnote ¹² For all their differences, the two kinds of scepticism mirrored one another, with more spiritual versions of physiology as their opposite pole.

Aligning ideology with Kant, against Kant's own French admirers, may also help to clarify the relationship between ideology and Rousseau, with the more strictly philosophical thought of Emmanuel-Joseph Sieyès forming a possible key to a fuller understanding of both alignments. Here, too, the parallel does not have to involve any real conceptual convergence, but a common aversion to another type of monism, in this case the monism underlying Claude-Adrien Helvétius's famous assertion that feeling and judging were identical. Rousseau's hostility to the claim is well known, as too is Helvétius's reply to Rousseau in his posthumously published De l'Homme (both were also well known in ideologue circles).Footnote ¹³ In this context, rejecting Helvétius meant establishing a distance from a purely physical, utility-driven conception of the progress of civilization, just as rejecting Kant (at least as he was construed in France) meant establishing a distance from the vitalist emphasis on the growing subordination of the physical to the spiritual as a measure of the same progress. Ideology lay somewhere in between. As Destutt de Tracy put it, its point was to counter those who “gravely assert that it is not necessary to know one's intellectual faculties in order to reason well, and that man in society does not need to study the principles of the social art”.Footnote ¹⁴ In fact, he went on to claim, “we are entirely works of art, namely of our own work, and are now as dissimilar today from the man of nature, or the original form of our existence, as is an oak from an acorn, or a chicken from an egg”.Footnote ¹⁵ If most of the mental processes uncovered by analysis were habitual and imperceptible, this still meant that they had once been acquired. “No one”, Destutt wrote, “is ignorant of the fact that the more often we repeat the same movement, whatever it may be, the more easily and rapidly we are able to perform it”:

The analogy applied both to thinking and to human improvement. “That much-vaunted golden age”, Destutt observed, “is an age of suffering and destitution, and the state of nature is a state of absolute stupidity and incapacity”. Its abiding aura was simply a trick played out by memory from one generation to the next, as the old looked back on their best years and passed on their recollections to the young.Footnote ¹⁷ Nothing, at the outset, separated human from animal life. In this sense, Destutt's characterization of humanity's beginnings was quite close to Rousseau's, but without Rousseau's emphasis on the solitary character of the state of nature or its bearing on his social contract theory. It was, instead, rather more like the picture of humanity's beginnings described in Condorcet's Esquisse d'un tableau historique des progrès de l'esprit humain (whose own intellectual debt to Rousseau has still to be worked out fully), but with a narrower analytical focus on the intellectual acquisitions involved in human improvement. These were susceptible to analysis, even if their underlying causes were not. More might become known (as Cabanis's researches promised to do), but ultimately the possibility of human improvement was explicable only in terms of the mysteries of the body's organization. That, Destutt emphasized, was all that could be said.Footnote ¹⁸ The concept of civilization was conspicuously absent from everything that Rousseau published, and it is most unlikely that this was an oversight. But, as will be shown in the next section, one approach to ideology is to describe it as an attempt to develop a theory of civilization that, notwithstanding Rousseau's analysis of modernity's pathology, was based on the premises of Rousseau's own criticism of modernity.

II

The reason for doing so is connected to the second of the terms with which the Décade philosophique is often associated, namely science sociale. As the Boulad-Ayoub edition shows, the term appeared quite frequently in the Décade but, as with the ideology–physiology pairing, it is still not clear what social science was supposed to be or do. Nor is it clear whether it was the same as, or different from, what, in opposition to those who claimed that reasoning and living in society simply came naturally, Destutt de Tracy called the social art. Usually, since science (meaning knowledge) is required for making something (meaning art), it looks as if social science was foundational and that social art was derivative, but this does not take the meaning of social science very far. The origins of the term are now quite well known.Footnote ¹⁹ So far, modern technology has not succeeded in unearthing any earlier occurrence of the phrase than the one published late in January 1789 in the first edition of Emmanuel-Joseph Sieyès's Qu'est-ce que le tiers état?. Sieyès, as is also well known, dropped the phrase from the two subsequent editions of the pamphlet that he published in 1789 and, in place of science sociale, referred instead to what he called la science de l'ordre social, or the science (or knowledge) of the social order, to support his claim that political decision-making in the forthcoming meeting of the French Estates General would have to be based on the principle of one member one vote, in order, as he put it, “to bind and engage the totality of representatives by one common will.” This, Sieyès wrote in the passage in which the neologism first appeared, was the form of decision-making that best fitted the principles of social science, or la science de l'ordre social. Those principles, he asserted, were “made [faits] to throw the most light” on the subject of political deliberation because they transcended claims based purely on particular interests, including those given simply by the size and composition of the French third estate.Footnote ²⁰ From this perspective, social science was an eminently political type of knowledge, with a distinctly moral orientation. It supplied the reasons underlying Sieyès's assertion that the last word in legislation did not belong to one or other of the various estates, houses or chambers that might make up any actual legislative assembly, or even to the numerical difference between the majorities and minorities involved in any particular legislative decision. It belonged instead, and in some more final sense, to the one common will that bound them all, independently, and also in some more final sense, to the various interests, parties or other types of grouping that might have been visible when the votes themselves were counted. On Sieyès's terms, social science was a metapolitical body of knowledge that justified the idea of majority rule. It did so too on loosely democratic premises because the majority in question was, for reasons that he did not specify in this particular passage, representative of the major part of the whole population. But quite why social science supplied this type of knowledge, and how it was supposed to work, were not questions that Sieyès addressed, at least in his published works. The nearest that he came to doing so, but without reference to the idea of a science sociale, was in the Preliminary to the Constitution, or An Acknowledgement and Explanatory Display of the Rights of Men and Citizens (as its 1795 English translation was entitled) that he presented to the French National Assembly's constitutional committee in July 1789 during its discussion of what was to become the Declaration of the Rights of Man. The conjectural history of needs and means that Sieyès outlined there came to form the starting point of a number of descriptions of social science when the subject became one of the responsibilities of the French Institute towards the end of 1795 and the Décade began its regular accounts of the institute's proceedings.

The institute's adoption of Sieyès's term did not mean that it had become self-explanatory. As the journal L'Historien, edited by the economist Pierre-Samuel Dupont de Nemours, reported in its issue of 10 December 1795, the decision to add the “analysis of sensations and ideas”, “morality” and “social science and political economy” to the established subjects of philosophy, history and geography had the effect of complicating the new institution's intellectual remit. “We confess”, L'Historien commented,

Although, the article noted, the institute's plan appeared to indicate that social science was intended to go alongside, rather than replace, legislation, all the branches of policy associated with political economy were also not exactly “alien” to the subject of legislation. Social science seemed, therefore, to be either unnecessary or misapplied. One way of dealing with this apparent confusion, the article continued, was to assume that

These uncertainties, it concluded, showed how difficult it was to classify human knowledge and how much more was still needed to clarify their relations.

Despite its reservations, the broad categories outlined in L'Historien's attempt at clarification were distinct enough. Social science, as Sieyès's original usage indicated, was to be the metadiscipline, with legislation (here meaning constitutional design) forming one of its derivative parts, and political economy (here meaning policy towards agriculture, industry, trade, public works, transport and taxation) forming a second, related, derivation. But, as the notice in L'Historien implied, this made the content of social science itself rather difficult to specify. It could not be legislation as such. Nor could it be legislation on the range of subjects (agriculture, manufacturing, trade, public works, navigation and taxation) that could be associated with political economy. If, as L'Historien suggested, social science was to have the same status as “the law of nature and nations”, then the parallel raised a further set of questions not only about why the new name was needed and what, by implication, was wrong with its predecessor, but also about whether the change of name amounted to a real change of content. This last question assumes, of course, that the content of “the law of nature and nations” was readily identifiable. But it is well known that, whatever its content was taken to be, the idea of the law of nature and nations was criticized very harshly in the eighteenth century, first by Jean-Jacques Rousseau, and then, towards the century's end, by Immanuel Kant. Kant's criticism can, here, be set to one side, not only because it was published several years after 1789, when Sieyès first coined the phrase science sociale, but also because Kant's famous characterization of Grotius, Hobbes and Pufendorf as “sorry comforters” followed the broad drift of Rousseau's own objections to the political thought of Hugo Grotius and his followers. If social science was the law of nature and nations under another name, it had, as will be shown, to be a version of the law of nature and nations that could accommodate Rousseau's criticism.

Rousseau's criticism of Grotius requires little rehearsal. “The science of political right [le droit politique]”, he wrote in a famous passage of his Emile in 1762, “is yet to be born and it is to be presumed that it never will be born”:

“The only individual”, Rousseau added immediately, “in a position to create this great and useless science was the illustrious Montesquieu”. But, he continued, Montesquieu himself was open to criticism. He was, Rousseau noted, “careful not to discuss the principles of political right. He was content to discuss the positive right of established governments, and nothing in the world is more different than these two studies.” Anyone, Rousseau wrote, wanting “to make healthy judgements about existing governments” had, unlike Montesquieu, to unite the two subjects, since it was “necessary to know what ought to be in order to judge soundly about what is”. This, however, gave rise to a further difficulty because it was not obvious that an individual would have an interest in doing either. Identifying the sources of the appropriate individual motivation and, by doing so, clarifying both “the principles of political right” and “the positive right of established governments” meant, Rousseau wrote at the end of this rather compressed set of assessments, finding answers to the questions “What importance does it have for me? and, What can I do about it?”Footnote ²²

The two questions that Rousseau raised can be taken to be an initial indication of the conceptual space that the idea of a social science was intended to fill. Both centred, in a very radical sense, on individual motivation and agency. They were questions about individuals, not families, communities or other types of nonpolitical association. Answering the first implied a version of “the principles of political right” (as Rousseau called them) that differed from those available in Grotius and Hobbes. Answering the second implied a continuation of Montesquieu's discussion of “the positive right of established governments”, but in a way that would remain compatible with the answer to the first question. The result, on these terms, would have been a non-Grotian version of “the principles of political right” that would fit whatever it was that Rousseau took to be significant in Montesquieu's discussion of “the positive right of established governments”. This description of social science is not particularly fanciful. Most contemporary evidence suggests that Sieyès's coinage was initially understood in terms that corresponded to Rousseau's anti-Grotian, qualified Montesquieuan, brief. But this still does not mean that it is particularly easy to describe what the substance of that brief might have been.

The fullest description of what this version of social science might have looked like had to wait until the second half of the nineteenth century before it appeared in print. This was the Cours d'organisation sociale, given, as its name indicates, as a public lecture course on a version of Sieyès's alternative name for social science by his political ally Pierre-Louis Roederer in the spring of 1793, but published only sixty years later as part of Roederer's collected works. The difference between “social order” and “social organization” may not look all that great, but the latter term was more redolent of physiology than the former (“organization” was a physiological term of art that referred to the complex structure of living beings and the variegated functions they housed). Although it is not clear whether Sieyès himself subscribed to whatever substituting “social organization” for “social order” may have meant, Roederer's main aim in the lecture course was to highlight the compatibility between the moral and political thought of Rousseau and Sieyès, against the uses to which Rousseau had been put both by Robespierre and, in different ways, by Jacques-Pierre Brissot and Tom Paine. As Roederer presented it, science sociale explained both why Rousseau's conception of government as an “elective aristocracy” was simply another name for what Sieyès called representative government and why a politics based on purely numerical majorities and minorities could still have binding political legitimacy.Footnote ²³ But the charged political circumstances in which the lecture course was given brought the explanation to a premature end.

In many respects, however, the strong emphasis that Roederer placed on the compatibility between the political thought of Rousseau and Sieyès had already been made in December 1791 in a pamphlet addressed to the former marquis de Condorcet that was published by the future French minister of the interior Dominique-Joseph Garat as a valedictory statement on the period he had spent reporting the debates in the National Assembly for the Journal de Paris between 1789 and 1791. Garat's aim in publishing the pamphlet was, he wrote, to encourage the members of the newly elected Legislative Assembly to follow a more principled and consistent line of conduct than their predecessors had done. His hope, in issuing a public invitation to someone with Condorcet's intellectual ability to take over his role, was that Condorcet's coverage of the debates in the new legislature would push its members to raise the quality of their arguments above the disappointingly low level at which, he claimed, they had been stuck throughout the lifetime of the National Assembly. The reason for this intellectual failure, Garat claimed, was that both what he called the “aristocratic” and “democratic” sides of the National Assembly had been unable to articulate the real principles underlying their respective positions. The aristocratic side had been largely content to appeal to precedent to justify its arguments in favour of preserving various aspects of the status quo. The democratic side had been equally content to assert its numerical majority and, when real argument was required, had relied instead on the largely self-evident character of its appeals to justice and equality, with the result that it had failed to identify a principled line of conduct when these values clashed (as had been the case, Garat noted, with the decision to confiscate the property of the Church, with the conflation of legislative and executive power in establishing the office of justice of the peace, and with the uncertainty surrounding the location of civil and military power among the assorted responsibilities of the newly created national guard, the gendarmerie and the regular army).

According to Garat, neither the democratic nor the aristocratic side had risen to the level of reasoned argument that the situation and opportunity required. On the aristocratic side, he suggested, the model that ought to have been followed was exemplified by the status and power of the Roman Senate. The ideal of disinterested but patrician rule that it had once embodied had, he wrote, been given its fullest theoretical expression in what he called

These “terrible sophisms”, in which “indisputable truths were mixed so artfully with harmful errors”, were the basis of the old idea that a wise minority equipped with theatrical rhetorical skills was the best means to establish “the rights of man and the principles of peace among nations”. This, Garat claimed, was the position represented by Hobbes. Social science, by implication, had something like the same objectives, but did not entail the negative judgement of “popular assemblies” made by “so many of the philosophers of the ancient and modern republics”. As Garat wrote in the same passage, this was the difference between Hobbes and Sieyès. If, he wrote, there had been a debate in the National Assembly between “a man like Hobbes on the one side and a man like the abbé Sieyès on the other” (or, he added, like Condorcet himself), the entitlements of the human race “after having been found again and proclaimed” would not have experienced the outrage to which they had been subjected “even by those who proclaimed them”.Footnote ²⁵ With a higher level of debate, he went on to assert, the National Assembly might even have gone beyond establishing a constitution, since “the political endeavour truly worthy of the progress of reason in this century is to remake societies after remaking constitutions, setting another goal for all those forces of the human race condemned otherwise simply to exhaust and lose themselves in seeking out mere scraps”.Footnote ²⁶

To reach this ambitious goal, Garat wrote at the end of the pamphlet, the new legislature would have to build on the theoretical achievements made by Sieyès in revising and correcting the political thought of Thomas Hobbes in the light of the conceptual innovations made by Montesquieu and Rousseau. The task, he emphasized, was more than a matter of some scholars doing a “little philosophy”, others doing “a little morality”, and yet others “a little physiology”. The kind of knowledge required would have to straddle them all. This meant that the new legislature would need the guidance supplied by something like the sort of common intellectual enterprise that an earlier generation had been given by Diderot and d'Alembert's Encyclopédie so that it could develop Rousseau's concept of sovereignty as the general will and Montesquieu's insistence that, as Garat put it, “powers have to be divided”. These, he wrote (borrowing the term that Sieyès had coined), “are the primary givens of science sociale”. But, he emphasized, they were still not social science itself. “With these ideas,” Garat wrote, “whose merit I am far from wanting to belittle, even nations which are held to be enlightened seem to have been left abandoned to the hazards of their passions and their faltering conceptions, just at the moment when they are constituting or reconstituting themselves.”Footnote ²⁷ A full version of social science, he continued, called for real understanding of the principles of taste and feeling underlying individual and collective judgement and discrimination. Broad principles, he insisted, were not enough. The “social art” would exist only when “men of profound and extensive minds, and with sensitive and brilliant imaginations present not only its principles, but plans of actions and models”. Here too the example of the ancients was telling. With Xenophon, there had been a model of a monarchy, just as, with Plato, there had been a model of a republic. Both, Garat commented, were writing for real people with real institutions. Their respective models simply gave “perfect expression” to what was already there. The need now, he continued, was for something analogous:

Garat's description of social science placed its intellectual origins in much the same context of early modern natural jurisprudence as L'Historien was to do, but it also made more explicit use of the broad parameters of Rousseau's earlier criticism of Grotius. It began with something like Rousseau's assertion that, notwithstanding his “sophisms”, Hobbes had identified the real content of natural law and, by doing so, had also shown that the underlying principles of Grotius's apparently more acceptable version were, in fact, identical. It continued with something similar to Rousseau's gesture towards Montesquieu, but narrowed it down to an endorsement of Montesquieu's emphasis on the need to divide political powers. Its most basic concept, however, was Rousseau's own redefinition of political sovereignty as the general will. This characterization of the theoretical origins of social science was not peculiar to Garat. Roederer, in his Cours d'organisation sociale of 1793, gave it something like the same intellectual starting point, while Condorcet, to whom Garat addressed his open letter, produced a theoretical map made up of much the same component parts. “Some philosophers like More and Hobbes”, he wrote in his posthumously published Sketch for a Historical Picture of the Progress of the Human Mind,

The passage reappeared (without acknowledgement of its source) in a pamphlet published in the same year, with the same somewhat surprising trilogy of Plato, More and Hobbes on one side, but with Montesquieu alongside Machiavelli on the other, and with Rousseau now occupying a position midway between the two.Footnote ³⁰ The two theoretical maps add a little more clarity to the idea of social science. Whatever it was, its content was not simply a set of general rules. These were already available in Plato, More or Hobbes and served to justify government by an enlightened elite, modelled on the class of guardians of Plato's Republic. Nor, as Rousseau's criticism of Montesquieu also indicated, was social science simply a product of careful observation of local norms and particular histories. It was a set of general rules that applied specifically to societies in which government was, in some sense, open to all, and in which norms and motives were, in both the first and last instance, radically individual in character.

This characterization of social science resurfaced in a review by the political economist Jean-Baptiste Say that was published in the December 1799–February 1800 issue of the Décade philosophique under the rubric science sociale. It was a review of a pamphlet by the medical philosopher Pierre-Jean-Georges Cabanis entitled Quelques Considérations sur l'organisation sociale en général et particulièrement sur la nouvelle constitution that was written, as its title indicated, to explain and justify the new republican constitution of the Year VIII in the light of a broader examination of what Cabanis, echoing the title of Roederer's Sieyès-inspired lecture course, called “social organization”. The allusion was deliberate. As both Cabanis in his pamphlet and Say in his review emphasized, the new constitution was designed to give France a second chance under the aegis of the political system that Sieyès had tried, but failed, to establish in 1789. The core of that system was its provision for all public offices to be subject to a combination of election and promotion, or “graduated promotion” as Sieyès's admirers called it, with eligibility for election to a particular office depending on a period of service in an office below. The result, Say wrote, would be a system of government able to give real existence to the “noble principles” proclaimed in 1789, because its starting point would no longer be given by cloudy speculations about a putative state of nature and putative natural rights. The new constitution took human needs and faculties as its starting point, not the state of nature or natural rights (Sieyès, as has been mentioned, had published an introduction to his draft declaration of the rights of man in 1789 that took the same course, but, like many of his proposals in that year, it soon fell by the political wayside).Footnote ³¹ As Say, following Cabanis, went on to claim, this more readily verifiable starting point made it easier to identify the qualities required by the type of government able to meet the specifications of the association in question. One, according to Cabanis, had to be strength, because a government had to be strong enough to protect individuals' liberty, but it also had to be skilfully weighted to ensure that it could not easily oppress those that it had to defend. A second quality was stability, but this too had to be combined with a capacity for every type of improvement. These qualities, as Cabanis had also indicated, all figured prominently in long-standing discussions of the assorted merits of democracy, aristocracy and monarchy, as well as combinations of the three. The new French constitution, Say claimed, combined the advantages of them all. It did so because the moderns could now boast of “two great discoveries”, namely the separation of powers and the representative system. The latter, he added, “is the most important; it can even be assumed, given the great size of modern states, that the representative system will sooner or later prevail everywhere.” Its key component, he continued, was “that ingenious mode of election, the fruit of Sieyès's meditations, by which the will of the people starts at the same time from both the base and summit of the social pyramid, to meet midway and settle on men who are the most educated [instruits], reliable [probes], and best-intentioned.” It meant, he wrote, quoting a speech by Garat of 14 December 1799 officially recommending that the constitution of the Year VIII should be ratified by a plebiscite, that “with no one being excluded, it will be quite difficult for anyone to be wrongly chosen.” For Cabanis himself, in a passage that Say quoted, Sieyès's representative system was simply “good democracy” (bonne démocratie).Footnote ³²

III

In this, Sieyèsian, sense social science was a metapolitical body of knowledge that served to justify the idea of majority rule under conditions of democratic political sovereignty. Here too the link to Rousseau is clear. Rousseau produced a concise description of the type of society in which both democratic political sovereignty and majority rule applied in the open Letter to Christophe de Beaumont, Archbishop of Paris that he published in 1763, soon after his Emile had been condemned by the Church, and banned by the parlement of Paris. Its starting point, he wrote, and the “fundamental principle of all morality” was that “man is naturally good.”Footnote ³³ This natural goodness, he emphasized, was not moral goodness, but simple self-love, or amour-de-soi. It was, Rousseau wrote, “the only passion which is born with man” and was “in itself indifferent either to good or evil.” This, he added, meant that he had been able to explain the origins of evil in a way that was more coherent than the Christian story of the Fall (explaining how someone truly innocent, like Adam, could sin, Rousseau pointed out, would either have to be incoherent, or end up by turning God into the author of evil).Footnote ³⁴ Since “none of the vices imputed to the human heart are natural to it”, he had, he wrote, “traced, as it were, their genealogy”, to show how “mankind are become what they are”, despite their original natural goodness. This “genealogy” involved three stages or “states of mankind.”Footnote ³⁵ The first was presocial, and revolved around amour-de-soi. This feeling, Rousseau emphasized, was a compound passion, with “two principles”, not one.Footnote ³⁶ The first applied to the body and the satisfaction of its physical needs. But the second applied to the mind, and entailed a different type of pleasure from simple physical well-being. The type of intellectual satisfaction that this involved was the product of what Rousseau called “the love of order” which, he continued, “expanded and become active, is denominated conscience”.Footnote ³⁷ The first principle gave rise to the purely self-centred feeling of pity, but the second led to a more moral awareness of others.

The difficulty was to find a way to keep amour-de-soi and amour de l'ordre in balance. The human condition, as Rousseau described it, was the outcome of the double-bind built into these two aspects of love. To get conscience going there had to be some sort of knowledge of order. But to get knowledge going there had to be relations and comparisons. Relations and comparisons, however, gave rise to amour-propre, which set the self and its interests alongside conscience and amour-de-soi. The two could coexist in a kind of precarious equilibrium for as long as common knowledge could override the individual self. So long, Rousseau wrote, “as the opposition of their interests is less than the concurrence of their knowledge, men are essentially good.” This, he added, “is the second state of mankind”.Footnote ³⁸ But, he continued,

then conscience turned into “a mere empty word which mankind reciprocally make use of to deceive each other.” This, Rousseau wrote, was humanity's “third and last state.”Footnote ³⁹ Dealing with it involved accepting what was now in place. Once there were selves and self-interests, the only available way out was a form of association that could make the public interest coincide with self-interest. As Rousseau put it,

“True policy”, in conditions in which “all the particular interests of individuals interfere and clash against each other”, meant reversing the standard characterization of the relationship between private goods and public goods. Instead of trying to find ways to make the former fit the latter, the problem was now to find ways to make the public good fit the private good. The formulation echoed Rousseau's criticism of Montesquieu in Emile. As he had put it there, the key questions that any description of political right had to address were “What importance does it have for me? and, What can I do about it?” The questions mattered when the social division of labour meant that individuals were radically dependent on others for their own survival needs. In conditions of social interdependence, Rousseau now asserted in his Letter to Christophe de Beaumont, individual interests had to come first. This, he added, meant that “I am now beginning to talk a strange language, as little understood by the majority of readers as by your Lordship”.Footnote ⁴¹

In one sense, Rousseau's “strange language” fitted his characterization of Hobbes, not Grotius, as the most authentic guide to what, nonetheless, were the sophistries of political right. In humanity's “third state”, when self-interest and opinion made individuals “enemies from their birth”, Hobbes's assessment applied. But it was still humanity's third state. From this perspective, Sieyès's coinage “science sociale” had a very literal meaning. It was the knowledge required to make and keep humanity social, but in a way that fitted the focus on individual interest and motivation underlying Rousseau's criticism of Grotius, Hobbes and Montesquieu. And, since it was knowledge, it could not simply be the knowledge that was compatible with humanity's first two states. For Rousseau, there was no society in humanity's first state, and no knowledge beyond an individual's ephemeral experiences. In the society that existed in humanity's second state, the opposition between individual interests was “less than the concurrence of their knowledge”. The type of knowledge required by social science had to fit the opposite state of affairs. It had, on Rousseau's terms, to provide guidance when the opposition between individual interests and opinions was more powerful than common knowledge. “The people of antiquity”, he emphasized in his Letters from the Mountains of 1764, “are not proper models of modern policy”. In the republics of the ancient world, slavery allowed those who were free to govern themselves. But the modern division of labour required “other maxims” and other political arrangements. “Every office”, Rousseau continued, “should be rendered so much the more easy to discharge as it is a trouble which puts you to expense and in which you do not engage as volunteers”. In conditions of social interdependence, when individual survival needs were inextricably entangled, democratic political sovereignty was essential. But the same conditions also meant reducing the duties of government as much as possible. Politics might be built into modernity, but modernity had to make the demands of politics minimal. The minimum was now, however, unavoidable. Rousseau drove the point home by citing a passage from the Polish prince Stanislas Leszczynski's recently published Le Philosophe bienfaisant: “We must take our choice, said the beneficent philosopher, and those who cannot put up with the work [ceux qui ne peuvent supporter le travail] will find rest only in servitude.” Modernity, on Rousseau's terms, was living proof of the old Latin proverb in libertate labor, in servitute dolor, cited in Stanislas Leszczynski's text.Footnote ⁴²

Sieyès's coinage, science sociale, followed this logic. But the term soon acquired a less strongly political and constitutional set of connotations than those indicated by Garat in the line of conceptual descent that he traced from Hobbes to Montesquieu, Rousseau and Sieyès. Garat's own pronouncements can, here, be taken as a register of the ambiguity. As he emphasized in 1791, the concepts of the separation of powers and the general will were no more than the primary elements of social science, since “remaking constitutions” was no more than the first step towards “remaking societies”. A full version of social science called for more emphasis on the feelings and judgements involved in taste, discrimination and culture. “Among the aliments that maintain man's life”, he wrote in his public letter to Condorcet, “it has been his taste, not his intelligence and understanding, that made all the choices and discoveries”. It had been taste, not intelligence and understanding, that was responsible for the glory of Greece. Analysis had come later, beginning with the scepticism of Bacon and continuing with Locke. This, although Garat did not use the term, meant that social science required aesthetics as much as analysis. “Humanity's finest days,” he wrote,

The speech that Garat made in 1799 officially recommending a plebiscite to ratify the constitution of the Year VIII repeated the same idea. France, he said, was a country covered in the lava of its earlier revolutionary eruptions (laves révolutionnaires). But, as the examples of Vesuvius and Aetna showed, volcanic eruptions could leave a longer-term legacy of fertility. Just as with real volcanoes, Garat claimed, the energy released by the Revolution was now available to be applied to “the sciences, the arts, commerce, and to all the forms of work that maintain and multiply the creations and the benefits of social existence”. Soon, he announced, “fields will be covered with all the riches of nature; cities will be decorated with the splendour of enlightenment [lumières] and the magnificence of the arts”. This was the future that the new constitution offered and, Garat concluded, “it is now at hand”.Footnote ⁴⁴

Garat's emphasis on the relationship between social science and the moral purpose and patriotic energy produced by nearly a decade of domestic upheaval and foreign war was similar in style to a Discours sur la science sociale given to the French Institute late in February 1798 by Napoleon's future arch-chancellor Jean-Jacques-Régis Cambacérès.Footnote ⁴⁵ This less directly political focus, with its stress on the moral capabilities unleashed by the events of the Revolution itself, gave social science a broader, but not obviously incompatible, content. According to Cambacérès, social science was the means to secure both individual and public well-being (bonheur) and, as a report in the Décade on his speech noted, served “to prove that the real object of that science is the perfecting of civilization”.Footnote ⁴⁶ The two goals complemented one another because (as Cabanis was also to emphasize two years later) the subject matter of social science began with human needs and faculties, not with a state of nature and natural rights. Human needs and faculties, Cambacérès stated, gave rise to the arts (in the broad, eighteenth-century sense of the term) and they in turn gave rise to authority since, with the arts, people needed protection, sometimes even from themselves. With authority came laws and, to be effective, the laws had to chime with morality, “that imperious ruler [dominatrice] of the human race by means of hope and fear”. The “arts, laws and morality” were, therefore, “the principal means to civilization and the true elements of social science”. Each housed a number of further subcategories, making social science a compound of several different, but interrelated, subjects. The arts embraced the subjects of agriculture, industry and trade; the laws included constitutional law, the civil law and the law of nations; while the subject of morality encompassed the whole range of human emotions or passions. The purely constitutional aspects of social science were, however, less significant. The “art of advancing civilization by laws”, Cambacérès observed, “consists less in perfecting constitutional law than in perfecting the civil law”. Civilization, he continued, was not a function of “relations between constituted powers and individuals”, but of real individual and social qualities, which meant that improvements to constitutional law did not necessarily imply improvements to “the social state”.Footnote ⁴⁷

Despite this reservation, legislation was still part of the trilogy of subjects that gave social science its content. Political economy, by way of the arts, formed the bonds of society; legislation maintained them by way of the constituted powers; while morality confirmed them by way of the duties that it entailed. Together, they supplied the means to happiness, “the goals of both society and social science”. Political economy bound people by interest, legislation by authority, and morality by feeling, which, Cambacérès continued, meant that the first dealt with people's physical faculties, the second with their rights, and the third with their emotions. In this sense, social science was truly “the science of man”. Its core concern was to identify “the best use of an individual's faculties, rights and passions”, particularly in present circumstances. “Ever since”, Cambacérès stated in his final peroration, “men have been reunited in society, never perhaps have nations found themselves in an age in which social science and the perfecting of society require so much attention”. This, he announced, was why the present age really was “the moment of social science”.Footnote ⁴⁸

It may well have been. As the years passed, however, the uncertainty surrounding the content of social science became less pronounced. One example of this change can be found in a series of Lettres sur l'éducation published in 1825 by an obscure admirer of Napoleon Bonaparte named Charles-Jean-Baptiste Bonnin, who had earlier welcomed Bonaparte's escape from the island of Elba with a pamphlet entitled De la Révolution européenne predicting that the restored imperial regime would finally put an end to patriotism, “that particular virtue of nations”, with its propensity to turn each people into strangers to one another, and replace it by the more noble spirit of humanity, when nations “come to feel that they are no more than the members of a great family, governed by similar laws and institutions and guided by the same spirit”.Footnote ⁴⁹ Bonnin set out the theoretical underpinnings of the message in his later work. The reason, he wrote there, that so many books on morality and politics went wrong was that they took “morality and politics to be two distinct sciences” and ignored the fact that “humanity's needs, and the relationships born of those needs, are the sources of both the one and the other”. Joining morality to politics, he continued, amounted to science sociale:

It was, it could be said, the law of nature and nations, minus the state-centred modern natural jurisprudence of Grotius and Hobbes. As Bonnin emphasized (here citing Condorcet), its principles applied to women as much as to men. Its substance, he also emphasized, was not a recent development, but stretched back to the political theology of the late seventeenth-century archbishop of Cambrai, François de Salignac de la Mothe-Fénelon, and the largely meritocratic social arrangements underpinning the heterodox doctrine of pure love or quietism with which Fénelon was associated. Initiation into the principles of political morality, Bonnin wrote, was available in “Fénelon's divine Dialogues of the Dead”. Others were also to be found in the same author's Telemachus, “a work of sublime reason”, in Montesquieu's Considerations on the Romans and “several books” of his “immortal” Sprit of Laws, as well as in Mably's Phocion's Conversations.Footnote ⁵¹ It was as if the French Revolution had never happened.

This, however, was not quite the case. Later in the same work, Bonnin, like Cabanis and Say a generation earlier, singled out Sieyès's introduction to his draft declaration of the rights of man of 1789 as the text which, he asserted, best embodied the principles of social science. But he also had no hesitation in reinforcing Sieyès's text by giving it a further set of articles dealing with the conditional nature of the right to private property which he took over (without mentioning his name) from Robespierre's draft declaration of the rights of man of 1793.Footnote ⁵² From this perspective, joining morality to politics meant fusing Sieyès's concern with needs and means with Robespierre's concern with property and community. In this new guise, social science no longer looked back to Hobbes but forward to the revolutions of 1848. The mixture of continuity and discontinuity built into this idea of social science, with Fénelon displacing Hobbes as the putative source of its original principles, also has a broader historiographical significance. It helps, in the first instance, to indicate how little, in intellectual terms, the French Revolution changed. In part, of course, this was because of how little, from the vantage point of 1825, the French Revolution really did seem to have changed. But it was also because it was genuinely not clear (not only in 1825) what, if anything, lay beyond what Rousseau had described as humanity's “third and last state”. One possibility was the increasing subordination of the physical to the spiritual posited by the type of vitalist natural philosophy that formed part of the intellectual context from which, or against which, ideology emerged. This, in the second instance, may help not only to make the broader intellectual continuities from the eighteenth to the nineteenth century easier to see, but also to get more of a grip on why Rousseau and the ambiguities in his thought were so pivotal in the intellectual life of both. The story of how Sieyès's phrase science sociale turned into the social science of Auguste Comte and Emile Durkheim has not yet been told. Nor is it clear how the related subjects of ideas, utility and civilization that, for a time, were held together precariously by ideology came to be overwritten by idealism and utilitarianism. Telling the whole story may mean going back to Leibniz and Fénelon as much as to Hobbes and Montesquieu, with Rousseau's “strange language” forming the point at which both the political and the largely nonpolitical (or, arguably, Hobbesian and Fénelonian) versions of social science began to take shape. In one guise, Rousseau's evaluation of the effects of the social division of labour pointed towards the never fully published political theories of Sieyès and Adam Smith. In another, as Charles-Jean-Baptiste Bonnin's now forgotten text indicates, it pointed towards the early nineteenth century's re-evaluation of both Robespierre and Rousseau together with the more recognizably Christian (or Jacobin) version of social science that became so prominent a feature of the socialism of 1848.Footnote ⁵³ Thanks to Josiane Boulad-Ayoub's edition of the Décade philosophique it may start to be easier to fill in some of the bits that lay in between, not only in France, and not only during the period of the French Revolution.